🛠 Tech Stack

💼 About This Role

You'll own the full post-training pipeline for Lovable's AI models, translating research into production training recipes for code generation and agent workloads. You'll build evaluation infrastructure and operate distributed training systems to get improvements to users in days or weeks. This role offers a chance to shape the AI behind a fast-growing product used by over 2 million people.

🎯 What You'll Do

• Own post-training pipeline from data curation to deployment
• Apply RL, preference optimization, SFT for code generation
• Build evaluation infrastructure for helpfulness, safety, latency
• Operate GPU orchestration and data pipelines at scale

📋 Requirements

• Run post-training jobs on large language models (RFT/RLVR, preference optimization)
• Write solid production code
• Fluent in PyTorch or JAX with distributed training
• Understand math behind preference optimization, reward modeling

✨ Nice to Have

• Worked on code generation or agentic use cases
• Put post-trained models into production with real users
• Owned full loop: data, training, eval, deployment, monitoring

🎁 Benefits & Perks

• 🚀 Extreme ownership and high velocity culture
• 🏢 Stockholm office with small, talent-dense team
• 🌍 Define the future of software creation

📨 Hiring Process

Estimated timeline: 2-4 weeks · AI estimate

1. 1Recruiter Call· 30 min
2. 2Technical Interview· 60 min
3. 3Team Interview· 45 min